A 'Retro-Inverso' PNA: Structural implications for DNA and RNA binding

Article abstract of DOI:10.1016/S0968-0896(98)00152-7

'Retro-inverso' peptide nucleic acid (PNA) monomers of thymine (T*: N-(amidomethyl)-N-(N¹-thyminylacetyl)-β-alanyl) (and adenine) have been prepared and introduced in PNA oligomers. A homo 'retro-inverso' T*₈ PNA was found not to hybridize to a complementary DNA or RNA oligonucleotide, whereas introduction of one retro-inverso thymine unit into the middle of a normal PNA 15-mer resulted in a ca. 8°C destabilization of the complex of this oligomer with a complementary DNA or RNA oligomer. In an effort to compensate for the structural nucleobase 'phase-shift' caused by the T*monomer by also introducing a β-alanine monomer it is concluded that the effect of the T* backbone is -7°C when hybridizing to DNA and -4.5°C when hybridizing to RNA. Nonetheless, the T*unit shows good sequence discrimination comparable to that of normal PNA. Molecular dynamics simulations indicate an unfavourable conformation of the backbone amide carbonyl group resulting in reduced interaction with the aqueous medium and an 'electrostatic clash' with the carbonyl of the nucleobase linker. These results show that a simple inversion of an amide bond in the PNA backbone has a dramatic, and hardly predictable, effect on the DNA mimicking properties of the oligomer. Copyright (C) 1998 Elsevier Science Ltd.

Full text of DOI:10.1016/S0968-0896(98)00152-7

BIOORGANIC &
MEDICINAL
CHEMISTRY
Bioorganic & Medicinal Chemistry 6 (1998) 1983±1992
A `Retro±Inverso' PNA: Structural Implications for DNA and
RNA Binding
Achim H. Krotz,a,* Sine Larsen,c Ole Buchardt,a,y Magdalena Erikssonb
and Peter E. Nielsenb,{
aCenter for Biomolecular Recognition, Chemical Laboratory II, The H. C. érsted Institute, University of Copenhagen, Universitetsparken 5,
DK-2100 Copenhagen é, Denmark
bCenterfor Biomolecular Recognition, IMBG, Department. of Biochemistry B, The Panum Institute, Blegdamsvej3, DK-2200 Copenhagen N
cCentre for Crystallographic Studies, Department of Chemistry, University of Copenhagen, Universitetsparken 5, DK-2100 Copenhagen é,
Denmark
Received 5 March 1998; accepted 21 May 1998
AbstractÐ`Retro±inverso' peptide nucleic acid (PNA) monomers of thymine (T*: N-(amidomethyl)-N-(N1-thyminyl-
acetyl)-b-alanyl) (and adenine) have been prepared and introduced in PNA oligomers. A homo `retro±inverso' T*8 PNA
was found not to hybridize to a complementary DNA or RNA oligonucleotide, whereas introduction of one retro±
inverso thymine unit into the middle of a normal PNA 15-mer resulted in a ca. 8 ꢀC destabilization of the complex of
this oligomer with a complementary DNA or RNA oligomer. In an e€ort to compensate for the structural nucleobase
`phase-shift' caused by the T* monomer by also introducing a b-alanine monomer it is concluded that the e€ect of the
T* backbone is 7 ꢀC when hybridizing to DNA and 4.5 ꢀC when hybridizing to RNA. Nonetheless, the T* unit
shows good sequence discrimination comparable to that of normal PNA. Molecular dynamics simulations indicate an
unfavourable conformation of the backbone amide carbonyl group resulting in reduced interaction with the aqueous
medium and an `electrostatic clash' with the carbonyl of the nucleobase linker. These results show that a simple
inversion of an amide bond in the PNA backbone has a dramatic, and hardly predictable, e€ect on the DNA
mimicking properties of the oligomer. # 1998 Published by Elsevier Science Ltd. All rights reserved.
Introduction
described the remarkable hybridization properties of
peptide nucleic acids (PNA) A (Scheme 1), a novel class
of nucleic acid binding DNA analogues with an achiral
and non-charged pseudo-peptide backbone consisting of
N-(2-aminoethyl) glycinyl units to which the nucleo-
bases are attached via methylenecarbonyl linkers.5±11
PNA oligomers containing the four natural nucleobases
are conveniently synthesized from monomeric building
units by Merri®eld solid-phase methodology.10±12 Mixed
purine±pyrimidine PNAs hybridize to complementary
DNA oligomers in a sequence-speci®c manner by for-
mation of double-helices obeying the Watson±Crick
base-pairing rules,9 whereas homopyrimidine PNAs
bind to complementary homopurine DNA sequences
with a 2:1 stoichiometry through (PNA)2/DNA triplex
formation by both Watson±Crick and Hoogsteen base-
pairing.13,14 Furthermore, such homopyrimidine PNAs
bind to double stranded DNA by duplex invasion.5,6
Molecules that bind mRNA or double-stranded DNA
with high anity and sequence speci®city represent
promising drug candidates in the antisense and antigene
therapy, and may also provide new tools in molecular
biology and diagnostics.1±4 In this context we recently
Key words: Nucleic acid; hybridization; antisense; peptide
nucleic acid.
*Present address: Isis Pharmaceuticals, 2292 Faraday Av.,
Carlsbad, CA 92008, USA.
yThe late.
{Corresponding author. Tel.: 45 35 32 77 62; fax: +45-
35396042; e-mail: pen@imbg.ku.dk
Abbreviations used: T: N-(amidoethyl)-N-(N1-thyminylacetyl)
glycinyl (n=2, m=1) b-ala-T: N-(amidoethyl)-N-(N1-thyminy-
lacetyl)-b-alanyl (n=2, m=2), T*: N-(amidomethyl)-N-(N1-
thyminylacetyl)-b-alanyl (n=1, m=2)
0968-0896/98/$ - see front matter # 1998 Published by Elsevier Science Ltd. All rights reserved
PII: S0968-0896(98)00152-7

1984
A. H. Krotz et al./Bioorg. Med. Chem. 6 (1998) 1983±1992
PNA oligomers with extended backbones (e.g. N-(ami-
doethyl) b-alaninyl) suggested that the interbase dis-
tance (six bonds along the backbone) in PNA A is best
suited for binding to a DNA target.22 The interchanging
of the positions of backbone methylene and side chain
carbonyl group has been found to abolish any interac-
tions with DNA,23 and reduction of the linker amide
likewise was found to be detrimental to the hybridiza-
tion properties.24 However, interchanging the position
of one carbonyl group of the backbone with a methy-
lene group from the backbone (`inverted amide', PNA
C) did not a€ect the Tm of PNA±DNA hybrids sig-
ni®cantly.25
As a logical extension of the work toward delineating
the structural elements which are responsible for high
binding anity of nucleic acid analogs, we have recently
designed a `retro±inverso' structural isomer (PNA B) of
the original aminoethyl-glycine (PNA A) (Scheme 1).26,27
In this PNA isomer two of the presumed key elements
present in the original aminoethyl glycine PNA back-
bone are retained; interbackbone distances (6 bonds
along the backbone and 3 bonds from the backbone to
the nucleobase) as well as restrained ¯exibility imposed
by the two `planar' amide groups.
We describe in this report in detail the convenient
synthesis of both a thymine and a N6-Cbz adenine
monomer suitable for the synthesis of `retro±inverso'
PNA B. Results from the ®rst X-ray crystallographic
study of a PNA monomer are also presented. Further-
more, the hybridization properties of di€erent oligomers
containing the new backbone modi®cation with DNA,
RNA and PNA oligomers are reported.
Results and Discussion
The four monomers of PNA A containing the natural
nucleobases are prepared in a straightforward fashion
from commercially available building blocks.11±13 The
retro-synthetic analysis of PNA B reveals a more chal-
lenging preparative task. The key structural feature of
monomers 4a and 4b (Scheme 2) is the inherently labile
N-(aminomethyl) amide moiety at the N terminus
(Scheme 3).27 As precursor we chose to prepare a pri-
mary amide, as the amide-to-amine conversion via
Hofmann rearrangement using Loudon's variant28,29
requires only very mild conditions. The synthetic route
to 4a and 4b is depicted in Scheme 2 as previously
reported.26,27
Scheme 1. Upper, peptide nucleic acid (PNA A) with the ori-
ginal N-(2-aminoethyl) glycine backbone; middle, `retro±
inverso' PNA B with the N-(aminomethyl) b-alanine backbone;
and lower, `inverted' PNA C. Note the di€erent positioning of
polar groups on the backbone, while the number of bonds
along the backbone between the nucleobases is kept constant.
B=nucleobase.
In vitro studies have demonstrated both good antisense
and antigene activity of PNAs.15±21
In order to understand the structural requirements for
high anity binding of nucleic acid analogs with
pseudo-peptide backbones more fully, we have engaged
in an investigation of the factors which may contribute
to the stability of PNA±DNA hybrids. Signi®cantly
lower thermostability (Tm) of hybrids of DNA with
X-ray crystallography
The crystal structure analysis revealed the presence of
two molecules in the asymmetric unit. As shown in

A. H. Krotz et al./Bioorg. Med. Chem. 6 (1998) 1983±1992
1985
Scheme 2. Synthesis of the N-protected monomers 5a and 5b. a. 2-chloroacetamide, K2CO3, DMF, 90 ꢀC, 24 h, 50%; b. BCH2COOH,
DCC/DhbtOH or HBTU, DMF/CH2Cl2 1:1, 0 ꢀC to r.t., 1 h (58±61%); c. I,I-[bis(tri¯uoroacetoxy)-iodo]benzene, CH3CN/H2O, r.t.,
20 h; d. di-tbutyl dicarbonate, K2CO3, dioxane, r.t., 1 h (c, d: 64±75%). e. LiOH, THF/H2O 1:1, r.t., 1 h (92%).
Scheme 3. Hydrolytic degradation of retro±inverso PNA.
Figure 1 they are almost related by non-crystallographic
inversion symmetry through the hydrogen bonds that
connect the two thymine rings. The almost planar con-
formation of the central amide group is also apparent.
The X-ray crystal structure analysis of 4a revealed a
conformation of the tertiary amide group as shown as
the major isomer (Scheme 4) found in solution.26
Hybridization Properties
The thermal stability of the PNA2±DNA triplexes
formed by T10PNA oligomers and the complementary
oligodeoxynucleotide (50-dA10-30, 50-dCGC-A10-CGC-
30) were determined (Table 1). The incorporation of a
single T* in position 5 of an oligomer A (II: H-T4-T*-T5-
LysNH2) resulted in a substantial decrease (48 ꢀC) of the
Tm (Tm=25 ꢀC) compared to the 10mer H-T10-LysNH2
(Tm=73 ꢀC). This large decrease in Tm is not totally
surprising since the important interbase distance
between the thymine of T* and the neighboring bases
is reduced to 5 bonds along the backbone to the N
terminal nucleobase and extended to 7 bonds to the C
terminal nucleobase, respectively, compared to 6 bonds
For the initial investigations of the hybridization prop-
erties of PNA with the new backbone modi®cation we
chose three types of poly-T oligomers: N-(aminoethyl)
glycine PNA with one modi®cation T* (II), N-(amino-
ethyl) glycine PNA with one T* and one b-ala-T mono-
mer at the N terminus of T* (IV) and a polyT* PNA
(VI) (Scheme 5).27

1986
A. H. Krotz et al./Bioorg. Med. Chem. 6 (1998) 1983±1992
in PNA A (Scheme 6). The melting temperature of II/
dA10 compares well with data obtained from PNA A H-
T5-LysNH2 (Tm=24 ꢀC). Thus the incorporation of one
backbone modi®ed monomer into an oligomer A leads
to a drastic distortion of the PNA A structure that pre-
vents cooperative binding of the PNA oligomer to its
complementary DNA. In an attempt to compensate
partly for the di€erent intraunit distances of the T*
monomer compared to the original N-(aminoethyl) gly-
cine backbone, we introduced an aminoethyl b-alanine
unit30,31 at the amino side of the T* unit. This resulted
in a 23 ꢀC increase in Tm (Tm=48 ꢀC), but the thermal
stability was still lower than that of the unmodi®ed
PNA by 23 ꢀC, and also lower by 11 ꢀC than that of the
PNA containing only the aminoethyl b-alanine unit
(Tm=59 ꢀC).31 However, from these experiments it is
not clear whether this decreased stability is to be ascri-
bed to the Watson±Crick or to the Hoogsteen strand of
the triplex, or both.
Consequently, we incorporated the PNA T* unit into
the middle of a 15-mer PNA (H-TGTACGT*CA-
CAACTA-NH2) that we have previously characterized
for an aminoethyl glycine PNA,9 and measured the
thermal stability of this PNA complexed with com-
plementary DNA, RNA or PNA oligomers. The results
(Table 2) showed a dramatic decrease (ÁTmꢁ8 ꢀC for
the DNA and RNA complexes) in thermal stability
upon introducing the T* unit. However, this unit also
causes the thymine to be `phase-shifted' 1 carbon on
the amino-side and +1 carbon on the carboxyl-side of
the nucleobase relative to the surrounding nucleobases
Figure 1. X-ray crystal structure of the `retro±inverso' mono-
mer. Drawing of molecule B and the more populated con-
former (0.68) of molecule A. The two molecules are virtually
mirror images of each other and di€er only slightly in a few
torsion angles: C6-N1-C7-C8 (A=100ꢀ.0(2) B= 99ꢀ.7(2), N1-
C7-C8-N9 (A=176ꢀ.8(2) B= 175ꢀ.3(2)), C7-C8-N9-C10
(A=179ꢀ.6(2) B= 179ꢀ.6(2)), C7-C8-N9-C20 (A=4ꢀ.1(3),
B= 3ꢀ.2(3)), N9-C20-N21-C22 (A=82ꢀ.4(3) B= 88ꢀ.2(3)).
Scheme 4. Rotamer equilibrium of PNA 4a (cf. ref26
)
Scheme 5. Retro±inverso monomers and oligomers.

A. H. Krotz et al./Bioorg. Med. Chem. 6 (1998) 1983±1992
1987
Table 1. Thermal stabilities of PNA-oligoT-DNA complexes
the complementary DNA or RNA oligonucleotide.
Thus there is little doubt that all three nucleobases are
engaged in conventional Watson±Crick base pairing
despite possible structural constraints imposed by the
changes in backbone structure.
PNA
Tm (ꢀC)
H-T10-LysNH2 (I)
H-T4-T*-T5-LysNH2 (II)
H-T5-LysNH2 (III)
73
25
24
48
59
H-T3-bT-T*-T5-LysNH2 (IV)
H-T4-bT-T5-LysNH2 (V)
Consequently, the retro±inverso backbone can be
accommodated in a PNA±DNA (RNA or PNA) duplex,
although at a considerable cost in duplex stability. Fur-
thermore, our results indicate that this backbone is
more easily accommodated in a PNA±RNA duplex than
in a PNA±DNA duplex and most easily in a PNA±PNA
duplex.
(Scheme 6). This could be the cause of the reduced
complex stability since it is known that introducing a
b-alanine backbone unit into a PNA oligomer (e.g.
PNA XII) (+1 carbon phase shift alone) results in a
comparable reduction in complex stability31 (Table 2,
column 6).
In agreement with the above results, we did not observe
any indication of complex formation in the range of 5 to
90 ꢀC between the fully `retro±inverso' 8-mer PNA B
(VI) and its complementary DNA oligomer; thereby
supporting the conclusion that the PNA B structure has
a low propensity of binding to DNA.
Therefore we attempted to compensate for the 1 frame
shift on the amino-side by introducing the b-alanine
backbone guanine (PNA IX). This causes a further
decrease (4 ꢀC) in complex stability (column 3), and
corresponds in terms of backbone structure to introdu-
cing one retro±inverso backbone (without phase shift)
plus a b-alanine backbone on the thymine. Thus we
should be able to indirectly deduce the e€ect of the
retro±inverso backbone within the context of the origi-
nal PNA glycine backbone by comparing PNA IX with
PNA X (Table 2, columns 3 and 4). The results show
ÁTm of 6 ꢀC and 4.5 ꢀC for the DNA and RNA com-
plexes and slightly less (3.5 ꢀC) for the PNA complex.
As a control, we also synthesized a PNA (PNA XI) in
which the b-alanine unit would not be compensatory.
As expected complexes with this PNA were less stable
(column 5).
Considering only geometric aspects, the orientation of
an amide group should have little impact on the hybri-
dization properties of PNA. However, reversal of the
amide groups in the backbone drastically reduces the
binding anity for complementary DNA sequences.
In order to better understand the results in structural
terms, we performed molecular dynamics simulations on
a (retro±inverso) PNA±DNA duplex. The solution
structure of the duplex 50dGACATAGC/H-GCTA
TGTC-NH2, previously determined by NMR meth-
ods,32 was modi®ed to contain only retro±inverso units
in the PNA strand by slipping the -CONH region of the
backbone on bond in the C-terminal direction, thereby
switching place with one of the methylene groups. The
modi®ed structure was subjected to molecular dynamics
(0.1 fs at 10 ꢀC), keeping the DNA strand essentially
®xed, followed by 104 steps of energy minimization. In
the resulting structure the most notable feature is the
direction of the carbonyl groups of the PNA backbone,
In order to analyze the integrity of the base pairs at and
around the positions of the modi®ed backbones, we
studied the e€ects of base pair mismatches at three
positions of the duplexes. All of the PNAs analyzed
show very similar relative responses to mismatches in
Scheme 6. `Phase-shift' e€ect of introducing a `retro±inverso' monomer (B) into a regular aminoethyl glycine (A) PNA oligomer. The
internucleobase `distances' are indicated above (5±7 bonds).

1988
A. H. Krotz et al./Bioorg. Med. Chem. 6 (1998) 1983±1992
Table 2. Thermal stabilities (Tm, ÁTm; ꢀC)a of PNA±NA complexes PNA: H-TGT AC[G TC]A CAA CTA-NH2 DNA/RNA: 30-
ACA TG[C AG]T GTT GAT-50
PNA VII
[GTC]
PNA VIIIc
[G(T*)C]
PNA IXb
PNA X
[G(bT)C]
PNA XI
[G(T*)(bC]
PNA XII
[(bG)TC]
[(bG)(T*)C]
Tm (ÁTm)
30-[CAG]
DNA
68,5 (0)
72,5 (0)
83,0 (0)
60,0 (8.5)
65,0 (7.5)
78,5 (4.5)
56,5 (12.0)
61,5 (11.5)
75,5 (7.5)
63,5 (5.0)
66,0 (6.5)
79,0 (4.0)
50,0 (18.5)
58,0 (14.5)
73,5 (9.5)
61,0 (7.5)
66,5 (6.0)
79,5 (3.5)
RNA
PNA2
ÁTm
30-[CTG]
DNA
RNA
10.0
12.5
8.0
10.0
6.5
9.0
11.0
10.5
5.5
11.5
10.0
11.0
30-[CCG]
DNA
RNA
12.5
11.5
8.0
10.0
13.5
9.5
11.5
11.0
5.5
11.0
11.0
12.0
30-[CGG]
DNA
RNA
10.0
7.5
7.5
8.5
6.5
5.5
9.0
6.0
5.5
9.0
8.0
7.5
30-[CAC]
DNA
RNA
19.5
18.0
16.5
16.5
15.0
14.5
19.0
15.0
8.0
13.5
17.0
18.0
30-[AAG]
DNA
RNA
18.0
16.0
15.0
15.5
6.5
13.5
16.0
16.0
ꢂ15
15.0
15.0
16.5
aBu€er: 100 mM NaCl, 10 mM Na-phosphate, 0.1 mM EDTA, pH 7.0.
bPNA: H-TAG TTG TGA CGT ACA-NH2.
cT*: `retro±inverso' thymine PNA unit.
dbG: b-alanine PNA unit.
which in most residues point into the helix, instead of
being directed outwards into solution as in the regular
PNA complexes.14,32±34 This could have destabilizing
e€ects on the (retro±inverso) PNA±DNA duplex for at
least two reasons. First, the carbonyl group of the
backbone is relatively close to the carbonyl group of the
backbone±nucleobase linker of the same residue, which
would be electrostatically unfavorable. Second, such
relocation of the backbone carbonyl will result in reduced
contact with the aqueous solvent which should likewise
cause destabilization. On the other hand, one of the mid-
dle backbone units in the crystal structure of a PNA±PNA
hexamer duplex indeed shows 50% `carbonyl in' struc-
ture,34 indicating that this conformation is not dramati-
cally destabilized. Therefore, the modeling does not o€er
an obvious explanation for the poor hybridization prop-
erties of the `retro±inverso' PNA backbone.
interbackbone distances has a dramatic e€ect on the
ability of the PNA to bind to complementary DNA or
RNA oligonucleotides and thus to mimic DNA. The
present results combined with those reported for
numerous previously described `peptide nucleic acids'
clearly show that the structural window that allows for
good DNA/RNA mimicking properties of a PNA is
very narrow indeed.
Experimental
Synthesis
2-Chloroacetamide, b-alanine ethyl ester hydrochloride,
3-hydroxy-1,2,3-benzotriazin-4(3H)-one (DhbtOH), 1,3-
dicyclohexylcarbodiimide (DCC), [bis(tri¯uoroacetoxy)-
iodo]benzene, di-tbutyl dicarbonate and tri¯uoroacetic
acid (TFA) were purchased from Aldrich. O-(lH-benzo-
triazolyl) N,N,N0,N0-tetramethyluronium hexa¯uoropho-
sphate (HBTU) was purchased from Nova. Bu€ers: pH
0: (HCl), pH 4: (56 mM citric acid, 68 mM NaOH,
44 mM NaCl, Fluka), pH 7 (26 mM KH2PO4, 41 mM
In conclusion we have shown that a simple inversion of
the amide in the PNA backbone structure which in
principle has conserved both the constrained ¯exibility
in terms of two amide groups (one in the backbone itself
and one in the linker to the nucleobase) as well as all

A. H. Krotz et al./Bioorg. Med. Chem. 6 (1998) 1983±1992
1989
Na2HPO4, Fluka), pH 9 (13 mM sodium tetraborate,
4.6 mM HCl, Fluka). Nl-Thyminyl acetic acid and N6-
(benzyloxycarbonyl)-adenin-9-yl acetic acid were pre-
pared as described.2 Melting points (uncorrected) were
determined in open capillaries on a Buchi melting
point apparatus (acc. Dr Tottoli) and are uncorrected.
lH and 13C NMR spectra were recorded on a Bruker
(250 MHz) or Varian spectrometer (400 MHz), respec-
tively. Compounds 2, 3 and 4 are mixtures of two rota-
mers. The resonances of the major isomer of 3a, 3b, 4a
and 4b are labeled with an asterisk. T=Thymine,
A=Adenine. High performance liquid chromatography
(HPLC) was carried out with a Waters 600E System
2H, CH2), 4.10 (q, J=7.1 Hz, 2H, OCH2), 7.53, 7.96 (2s,
2H, NH2), 9.22 (s, 2H, NH2+) ppm. 13C NMR
([D6]DMSO): d=14.0 (CH3), 30.2 (CH2C(O)), 42.4
(NCH2CH2), 47.6 (NCH2C(O)), 60.5 (OCH2), 166.8
(amide CˆO), 170.0 (ester CˆO) ppm. C7H15ClN2O3
(210.66): calcd C 39.91, H 7.18, N 13.30; found C 39.45,
H 7.10, N 13.22.
N-(Carbamoylmethyl)-N-(thymin-1-ylacetyl)-ꢀ-alanine
ethylester (2a). DhbtOH (9.37 g, 57.4 mmol), N1-thymi-
nyl acetic acid (10.57 g, 57.4 mmol) and CH2Cl2
(100 mL) were added to
a solution of 1 (10.0 g,
57.4mmol) in dry DMF (100 ml). The solution was cooled
to 0 ꢀC in an ice bath and a solution of DCC (1.90 g,
9.18mmol) in DMF:CH2Cl2 (100 ml, 1:1, v/v) was added
dropwise within 20 min. The ice bath was removed and
after stirring for 1 h at r.t. the reaction mixture was ®l-
tered, the precipitate (DCU) was rinsed with CH2Cl2
(50 mL) and the organic solutions and CH2Cl2 (450 mL)
were combined. A colorless precipitate formed overnight.
Filtration, rinsing with CH2Cl2 and drying in vacuo
a€orded 2a (12.0g, yield 61%), mp 207±208 ꢀC.
Controller, equipped with
a Waters 486 Tunable
Absorbance Detector (detector wavelength: 260 nm) and
a Waters Data Module. We used a C18-reversed phase
column from Vydag. The chromatographic conditions
were as follows: A: from CH3CN:H2O:TFA 1:99:0.1
1
to 10:90:0.1 in 20 min, ¯ow rate=1.5 mL min
to=2.3 min. B: from CH3CN:H2O:TFA 1:99:0.1 to
,
10:90:0.1 in 10 min, then to 70:30:0.1 in 30 min, ¯ow
1
rate=1.5 mL min
. Fast-atom bombardment mass
spectra (FAB-MS) were taken on a Masslab VG 12-250
quadrupole instrument ®tted with a VG-FAB source
and a JMS-Hx/Hx 110A High Performance Tandem
Mass Spectrometer (JEOL). Peaks at higher m/z can be
assigned to [M+Li]+ or [M+Na]+. Micro analyses
were performed at the H. C. érsted Institute.
Due to restricted conformations at the tertiary amide
group two sets of signals (ratio 1:1) are observed for
most protons and carbons in the NMR spectra, the
coalescence temperature is 85 ꢀC. 1H NMR
([D6]DMSO, r.t.): d=1.16, 1.18 (2t, J=7.1 Hz, 3H,
CH2CH3), 1.73 (s, 3H, T-CH3), 2.49, 2.70, 3.45, 3.58 (4t,
J=7 Hz, 4H, CH2CH2), 3.84, 4.04 (2s, 2H, CH2), 4.03,
4.07 (2q, J=7 Hz, 2H, OCH2), 4.42, 4.65 (2s, 2H, T(N1)-
CH2), 7.01, 7.22, 7.26, 7.54 (4s, 2H, NH2), 7.28, 7.32 (2s,
1H, CH), 11.22, 11.24 (2s, T, NH) ppm. 13C NMR
([D6]DMSO): d=11.92, 11.95 (T, CH3), 14.1 (CH2CH3),
32.1, 32.7 (CH2CO), 43.4, 43.9 (NCH2CH2), 47.85,
47.91 (NCCH2N), 48.5, 50.2 (T(N1)-CH2), 60.1, 60.2
(OCH2), 108.0, 108.1 (CCH3), 142.2, 142.3 (CH), 151.0,
151.1 (T, NC(O)N), 164.4 (T, C ˆ O), 167.1, 167.7 (tert.
amide C ˆ O), 169.8, 170.0 (prim. amide C ˆ O), 171.2
(ester C ˆ O) ppm. FAB-MS: m/z=341.1 (M+1)+.
C14H20N4O6 (340.34): calcd C 49.41, H 5.92, N 16.46;
found C 48.76, H 5.76, N 16.34.
N-(Carbamoylmethyl)-ꢀ-alanine ethylester (1). A sus-
pension of 2-chloroacetamide (2.34 g, 25.0 mmol), b-
alanine ethyl ester hydrochloride (15.36 g, 100 mmol)
and K2CO3 (42.0 g) in dry DMF (200 mL) was heated at
90 ꢀC for 24 h. After cooling to r.t. the reaction mixture
was ®ltered and the solvent and unreacted b-alanine
ethyl ester were removed by distillation under reduced
pressure (50 ꢀC, 1 Torr). A yellow oil (6.0 g) remained.
Pure 1 (2.24 g, 50%) was obtained as colorless oil by
¯ash chromatography (silica 20Â5 cm, 1. ethyl acetate,
2. ethyl acetate:ethanol 2:1, 3. ethanol). TLC (methanol,
det. ninhydrin), Rf(1)=0.65. 1H NMR ([D6]DMSO):
d=1.18 (t, J=7.1 Hz, 3H, CH3), 2.42, 2.70 (2t,
J=6.7 Hz, 4H, CH2CH2), 3.03 (s, 2H, CH2), 4.06 (q,
J=7.1 Hz, 2H, OCH2), 7.02, 7.24 (2s (br), 2H, NH2)
ppm. 13C NMR ([D6]DMSO): d=14.1 (CH3), 34.5
(CH2CO), 44.7 (NCH2CH2), 51.8 (NCH2CN), 59.8
(OCH2), 172.1 (C ˆ O), 173.5 (C ˆ O) ppm. FAB-MS:
m/z=175.1 (M+1)+.
N-(Carbamoylmethyl)-N-(N6-(benzyloxycarbonyl)adenin-
9-ylacetyl)-ꢀ-alanine ethylester (2b). According to the
procedure for the preparation of 2b, 1 (2.00 g, 11.5 mmol)
was reacted with N9-(N6-benzyloxycarbonyl)adeninyl
acetic acid (3.76 g, 11.5 mmol), DhbtOH (1.88 g,
11.5 mmol) and DCC (2.37 g, 11.5 mmol). Flash chro-
matography on silica (twice, 25Â5 cm, ethyl acetate:
methanol 3:1) a€orded 2b (4.0 g, DCU content ca.
15%).
A solution of 1 (530 mg) in diethyl ether was treated
.
with HCl (4.7M in diethyl ether). The precipitate 1 HCl
was ®ltered, rinsed with cold diethyl ether and dried in
vacuo, white powder.
Using HBTU as the activating agent we obtained 2b of
.
higher purity. 1 HCl (211 mg, 1.0 mmol) was dissolved in
DMF (2.5 mL), pyridine (2.5 mL) and N-diisopropyl-
1H NMR ([D6]DMSO): d=1.21 (t, J=7.1 Hz, 3H,
CH3), 2.80, 3.16 (2t, J=7.5 Hz, 4H, CH2CH2), 3.69 (s,

1990
A. H. Krotz et al./Bioorg. Med. Chem. 6 (1998) 1983±1992
ethyl amine (0.64 mL). N9-(N6-benzyloxycarbonyl)-
adeninyl acetic acid (327 mg, 1.0 mmol) and HBTU
(380 mg, 1.0 mmol) were added subsequently. After stir-
ring for 1 h at r.t. toluene (30 mL) was added. The pre-
cipitate was isolated by ®ltration. After ¯ash
chromatography (silica, 8Â2 cm, eluent: methanol:ethyl
acetate 1:1) 2b (280 mg, yield 58%) was obtained as a
white amorphous solid, mp 202±203 ꢀC.
(tBu-CH3), 32.2 (CH2CO), 41.4 (NCH2CH2), 48.1
(T(N1)CH2), 53.1 (NCH2N), 60.0 (OCH2), 78.8
(C(CH3)3) 108.2 (T, CCH3), 142.0 (CH), 151.1(T,
NC(O)N), 155.7 (carbamate C ˆ O), 164.4 (T, C ˆ O),
167.0 (tert. amide C ˆ O), 171.1 (ester C ˆ O) ppm.
FAB-MS: m/z=413.20 (M+H)+. C18H28N4O7 (412.44)
calcd C 52.42, H 6.84, N 13.58; found C 52.37, H 6.68,
N 13.57.
1H NMR ([D6]DMSO): (two rotamers, ratio 1:1)
d=1.25, 1.30 (2t, J=7.1 Hz, 3H, CH3), 2.60 (overl. with
solvent), 2.89, 3.57, 3.81 (4t, J=7.1 Hz, 4H, CH2CH2),
3.96, 4.29 (2s, 2H, CH2), 4.12, 4.21 (2q, 2H, CH2CH3),
5.23, 5.48 (2s, 2H, CH2), 5.30 (s, 2H, CH2Ph), 7.12, 7.77
(2s, 1H, NH2), 7.40±7.55 (m, 6H, aromatic H, NH2),
8.41 (s, 1H, A-H), 8.67, 8.68 (2s, 1H, A-H), 10.71 (s, 1H,
HNCbz) ppm. 13C NMR ([D6]DMSO): (two rotamers)
d=14.08, 14.13 (CH3), 32.08, 32.64 (CH2CO), 43.51,
43.95, 44.12, 44.16 (CH2NCH2), 48.47, 50.29 (A(N9)
CH2), 60.05, 60.23 (OCH2CH3), 66.27 (OCH2Ph),
122.90, 122.96, 127.83, 127.96, 128.40, 136.40, 145.27,
145.32, 149.31, 151.43, 152.18, 152.40, 152.44, 166.44,
167.12 (tert. amide CˆO), 169.78, 169.83 (prim. amide
CˆO), 171.22 (ester CˆO) ppm. FAB-MS: m/z
N-(N-tert-Butyloxycarbonyl-aminomethyl)-N-(N9-benzyl-
oxycarbonyl-adenin-9-ylacetyl) ꢀ-alanine ethylester (3b).
A solution of amid 2b (2.60 g, 5.38 mmol) and I,I-[bis
(tri¯uoroacetoxy)iodo]benzene (2.77 g. 6.45 mmol) in
CH3CN/H2O (50 mL) was kept for 20 h at r.t.. The
mixture was ®ltered and the solution was evaporated to
dryness under reduced pressure. 1,4-Dioxane (50 mL)
and Na2CO3 (5.0 g) were added, followed by addition
of
a solution of di-tert-butyl dicarbonate (1.65 g,
7.56 mmol) in 1,4-dioxane (10 mL). The mixture was
stirred at r.t. for 24 h, ®ltered and 3b (2.24 g, yield 75%)
was obtained by ¯ash chromatography (silica, 20Â5 cm,
ethyl acetate:methanol 97:3). TLC (MeOH:ethyl acet-
ate 1:1, det. UV, ninhydrin): Rf(2b)=0.65, Rf(i)=0.47,
Rf(3b)=0.85. 1H NMR ([D6]DMSO): d=1.23*, 1.30
(2t, J=7.0 Hz, 3H, CH2CH3), 1.46, 1.52* (2s, 9H, tBu),
2.6* (overl. with solvent), 2.90 (2t, CH2CO2), 3.60*, 3.90
(2t, J=7.3 Hz, 2H, NCH2CH2), 4.11*, 4.20 (2q,
J=7.1 Hz, 2H, CH2CH3), 4.65, 4.82* (2d, 2H, NCH2N),
5.30 (s, 2H, CH2Ph), 5.41, 5.53* (2s, 2H, CH2-A), 7.40±
7.60 (m, 5H, aromatic H), 8.02 (t, NH), 8.37, 8.40, 8.66
(3s, A-H), 10.72 (s, NH) ppm. FAB-MS: m/z=556.4
(M+1)+. C26H33N7O7 (555.59) calcd C 56.21, H 5.99,
N 17.65; found C 55.65, H 5.89, N 17.16.
=484.12 (M+H+). C22H25N7O6 (483.48) calcd
C
54.65, H 5.21, N 20.28; found C 52.93, H 5.09, N 19.84.
N-(N-tert-Butyloxycarbonyl-aminomethyl)-N-(N1-thy-
minylacetyl) ꢀ-alanine ethylester (3a). A solution of I,I-
[bis(tri¯uoroacetoxy)iodo]benzene (16.71 g, 38.8 mmol)
in CH3CN (55 mL) was added dropwise within 10 min
to a stirred solution of 2a (11.5 g, 33.8 mmol) in
CH3CN:H2O (275 mL, 2:3 v/v). After 20 h at r.t. the
volume was reduced to ca. 50 mL under reduced pres-
sure (1 Torr, 30 ꢀC) and the solution was extracted with
diethyl ether (150 mL). The aqueous phase was evapo-
rated to dryness in vacuo (1 Torr, 30 ꢀC). The remaining
solid was rinsed with diethyl ether (20 mL) and dried in
vacuo. 1,4-Dioxane (150 mL), K2CO3 (12.2 g) and a
solution of di-tert-butyl dicarbonate (8.4 g, 38.5 mmol)
in 1,4-dioxane (50 mL) were added. After stirring for 1 h
at r.t. the reaction mixture was ®ltered and the solvent
was removed under reduced pressure. Puri®cation by
column chromatography (silica, 35Â10 cm, hexane:
acetone 1:1) and subsequent precipitation from MeOH
(60 mL) by addition of hexane (300 mL) a€orded 3a
(8.6 g, yield 64%), mp 174±175 ꢀC. TLC (MeOH, det.
UV, ninhydrin): Rf(2a)=0.73, Rf(i)=0.5, Rf(3a)=0.85.
1H NMR ([D6]DMSO) d=1.17 (t, J=7 Hz, 3H,
CH2CH3), 1.38, 1.42* (2s, 9H, tBu), 1.75 (s, 3H, T-
CH3), 2.50*, 2.72 (2t, J=7.9 Hz, 2H, NCH2CH2), 3.31*,
3.50 (2t, 2H, NCH2CH2), 4.05*, 4.08 (2q, 2H, OCH2),
4.58, 4.61* (2d, 2H, CH2), 4.61 (overlap.), 4.74* (s, 2H,
CH2), 7.26*, 7.35 (2s, 1H, CH), 7.4, 7.8 (2m, 1H, NH),
11.25 (s, 1H, NH) ppm. 13C NMR (major rotamer,
[D6]DMSO): d=11.9 (T, CH3), 14.1 (CH2CH3), 28.0
N-(N-tert-Butyloxycarbonyl-aminomethyl)-N-(N1-thy-
minylacetyl) ꢀ-alanine (4a). A solution of 3a (5.45 g,
13.2 mmol) in THF (80 ml) and aqu. LiOH (0.5 M,
80 mL) was stirred at r.t. for 1 h and extracted with ethyl
acetate. The aqueous layer was acidi®ed to pH 3.5 by
dropwise addition of hydrochloric acid (4 M) and kept
at 4 ꢀC for 3 h. The colorless precipitate was ®ltered,
rinsed with water and dried in vacuo to a€ord 4a
(4.40 g), mp 196 ꢀC (dec.). The ®ltrate was extracted
with ethyl acetate, the organic phase dried over Na2SO4
and the solvent was removed under reduced pressure.
The residue was crystallized from MeOH:ethyl acetate:
hexane to a€ord additional 4a (280 mg), total yield
92%. TLC (CHCl3:MeOH:NEt3 7:2:1, UV-detection):
Rf(3a)=0.94, Rf(4a)=0.52). 1H NMR ([D6]DMSO):
(two rotamers, ratio 3:1) d=1.46, 1.50* (2s, 9H, tBu),
1.83 (s, 3H, T-CH3), 2.52*, 2.73 (2t, J=7.5 Hz, 2H,
NCH2CH2), 3.54*, 3.64 (2t, J=7.3 Hz, 2H, NCH2CH2),
4.64, 4.69 (2d, J=6.4 Hz, J=7.0 Hz. 13C NMR (major
rotamer, [D6]DMSO): d=11.9 (T, CH3), 28.1 (tBu-
CH3), 32.2 (CH2CO), 41.6 (NCH2CH2), 48.1 (T(N1)
CH2), 53.2 (NCH2N), 60.0 (OCH2), 78.9 (C(CH3)3),

A. H. Krotz et al./Bioorg. Med. Chem. 6 (1998) 1983±1992
1991
108.2 (T, CCH3), 142.0 (CH), 151.1 (T, NC(O)N), 155.7
(carbamate C ˆ O), 164.4 (T, C ˆ O), 166.9 (tert. amide
C ˆ O), 172.7 (carboxylic acid C ˆ O) ppm. FAB-MS:
m/z=385.15 (M+H)+. C16H24N4O7 (384.39): calcd C
50.00, H 6.29, N 14.58; found C 49.95, H 6.40, N 14.23.
then added to the solid residue obtained under A. After
stirring for 2 h at room temperature the reaction mixture
was puri®ed by column chromatography on silica using
a methanol/ethyl acetate gradient as eluent to give the
oligomers.
N-(N-tert-Butyloxycarbonyl-aminomethyl)-N-(N6-benzyl-
oxycarbonyladenin-9-ylacetyl) ꢀ-alanine (4b). A mixture
of ester 3b (2.34 g, 4.21 mmol) in THF (25 ml) and aqu.
LiOH (0.5 M, 25 mL) was stirred at r.t. for 1.5 h, when
TLC indicated that hydrolysis was complete. Extraction
with CH2Cl2 (extract discarded), acidi®cation with HCl
(2N) and extraction with CH2Cl2 a€orded 4b (2.1 g,
95%). 1H NMR ([D6]DMSO-d6): d=1.47, 1.53* (2s,
9H, tBu), 2.51*, 2.78 (2t, J=7.5 Hz, 2H, NCH2CH2),
3.57*, 3.75 (2t, J=7.2 Hz, 2H, NCH2CH2), 4.65, 4.83*
(2d, NCH2N), 5.29 (s, 2H, CH2Ph), 5.44, 5.54* (2s, 2H,
CH2), 7.4±7.6 (5H, aromat.), 8.02 (NH), 8.39, 8.67 (2s,
aromat.), 10.7 ppm. FAB-MS: m/z=528.1 (M+1)+.
C24H29N7O7 (527.53): calcd C 54.64, H 5.54, N 18.59;
found C 54.86, H 5.50, N 18.81.
Hydrolysis, general procedure. C. In a typical experi-
ment, the oligomer-ester (0.1±1 mmol) was added to a
mixture of THF (0.8±8 mL) and LiOH (0.5 M, 0.8±
8 mL) and the mixture was stirred at room temperature
until TLC indicated that hydrolysis was complete (ca.
2 h). After extraction with n-butanol, the aqueous phase
was acidi®ed with HCl (2N). A precipitate was iso-
lated by ®ltration and/or the solution was extracted
with n-butanol. The organic layer was separated and
dried over Na2SO4, evaporated and the crude products
were combined and puri®ed by ¯ash chromatography
(silica) using
a methanol/ethyl acetate gradient to
give the oligomer-acid of high purity (>95%, acc. to
HPLC). The identity was con®rmed by mass spectro-
scopic analysis.
N,N-(ammoniummethyl)-(N1-thyminylacetyl)-ꢀ-alanine
tri¯uoroacetate (5a). At room temperature, 4a (102 mg,
0.265 mmol) was dissolved in 95% TFA/cresol (1 mL).
After 5 min the solution was evaporated in vacuo and
the residue was dissolved in ethanol (0.5 mL). Upon
addition of diethyl ether a precipitate formed (75 mg).
FAB-MS: m/z=285.1 (M+).
X-ray crystallography. Crystals were obtained by slow
evaporation from a [D6]DMSO solution. Crystal size
0.15Â0.30Â0.60 mm3, monoclinic, space group P21/n,
T=122K, a=9.9263(13), b=19.098(5), c=22.043(4)
A, b=93.786(12)ꢀ, V=4169.6(14) A3, Z=8, dcal
=1.311 gcm 3, CAD4 di€ractometer, CuKa radiation,
2!yꢃ140ꢀ.9698 measured re¯ections, 7893 unique, cor-
rections for background, Lorentz, polarization and
decay with DREAD programs.35 Structure solved by
direct methods SHELXS-86, full matrix least squares
re®nement on /F/2 SHELXL-93.36 The ethyl ester group
of one of the independent molecules is disordered, it can
be described by two di€erently populated (0.68(4) and
0.32(3)) conformers. The disordered carbon atoms were
given isotropic, all other non-hydrogen atoms aniso-
tropic displacement parameters. Hydrogen atoms were
located in the di€erence map apart from those attached
to the disordered ethyl group. Their positional para-
meters were not re®ned, displacement parameters for
CH3 hydrogen atoms are 1.5ÂUiso of the atoms to
which they are bonded, 1.2ÂUiso for all other hydrogen
atoms. wR2=0.204 R1=0.0668 for re¯ections with/Fo/
>4s (/Fo/). Further details of the crystal structure
investigation are available on request from the Director
of the Cambridge Crystallographic Data Centre, 12
Union Road, GB-Cambridge CB2 1 EZ (UK) on quot-
ing the full journal citation.
N-(N1-thyminylacetyl)-ꢀ-alanine (6). B. A suspension of
4a (100 mg, 0.26 mmol) and HCl/1,4-dioxane (0.5 M)
was heated for 5 min at 100 ꢀC. Evaporation and ¯ash
chromatography of the residue on silica a€orded 4 in
nearly quantitative yield. 1H NMR ([D6]DMSO):
d=1.82 (s, 3H, CH3), 2.47 (t, J=6.8 Hz, 2H,
NCH2CH2), 3.34 (m, 2H, HNCH2), 4.33 (s, 2H, T-
CH2), 7.49 (s, 1H, CH), 8.27 (t, J=5.4 Hz, 1H,
HNC(O)), 11.3 (s, 1H, T-NH), 12.3 (s, (br), 1H, OH)
ppm. 13C NMR ([D6]DMSO): d=11.9 (CH3), 33.8, 34.9
(CH2CH2), 49.3 (T-CH2), 107.9 (CCH3), 142.3 (CH),
151.0, 164.4, 166.9, 172.7 (4C ˆ O) ppm. FAB-MS: m/z
=256.1 (M+1)+.
Peptide coupling; general procedures. A. In a typical
experiment, amino acid ester (0.1±1 mmol) was dis-
solved in tri¯uoroacetic acid:CH2Cl2 (0.2±2 ml, 1:1, v/v).
After stirring for 20 min at room temperature the clear
solution was evaporated to dryness in vacuo (0.5 Torr).
Diethylether (2 mL) was added and the mixture was
evaporated to dryness.
Acknowledgements
B. Equimolar amounts of tBoc-protected amino acid
(0.1±1 mmol) and HBTU (0.1±1 mmol) were dissolved in
a solution of DMF:pyridine (0.5±5 mL, 1:1, v/v) and
diisopropylethyl amine (0.2±2 mL). This solution was
This work was supported by the Danish National
Research Foundation and a DANVIS fellowship to
AHK. We thank Inga Nordstrùm, Flemming Hansen

1992
A. H. Krotz et al./Bioorg. Med. Chem. 6 (1998) 1983±1992
and Annette W. Jùrgensen for skilful technical assis-
tance.
M. A.; Au, K. G.; Carter, S. G.; Bruckenstein, D. A.; Boyd, A.
L.; Noble, S. A.; Babiss, L. E. Science 1992, 258, 1481.
18. Knudsen, H.; Nielsen, P. E. Nucleic Acids Res., 1996, 24,
494.
References and Notes
19. Vickers, T. A.; Grith, M. C.; Ramasamy, K.; Risen, L.
M.; Freier, S. M. Nucleic Acids Res., 1995, 23, 3003.
1. Crooke, S. T. In Burger's Medical Chemistry and Drug Dis-
covery; Wol€, M. E, Ed.; John Wiley & Sons, Inc.: 1995, pp
863±900.
20. Bonham, M. A.; Brown, S.; Boyd, A. L.; Brown, P. H.;
Bruckenstein, D. A.; Hanvey, J. C.; Thomson, S. A.; Pipe, A.;
Hassman, F.; Bisi, J. E.; Froehler, B. C.; Matteucci, M. D.;
Wagner, R. W.; Noble, S. A.; Babiss, L. E. Nucleic Acids Res.,
1995, 23, 1197.
2. Uhlmann, E.; Peyman, A. Chem. Rev. 1990, 90, 544±584.
3. De Mesmaeker, A.; Altmann, K. H.; Waldner, A.; Wende-
born, S. Current Opinion in Structural Biology 1996, 5, 343.
4. Matteucci, M.; Wagner, R. W. Science 1996, 384(supp) 20.
21. Good, L.; Nielsen, P. E. Antisense and Nucleic Acid Drug
Development 1997, 7, 431.
5. Nielsen, P. E.; Egholm, M.; Berg, R. H.; Buchardt, O. Sci-
ence 1991, 254, 1497.
22. Hyrup, B.; Egholm, M.; Rolland, M.; Nielsen, P. E.; Berg,
R. H.; Buchardt, O. J. Chem. Soc. Chem. Commun. 1993, 518.
6. Cherny, D. Y.; Belotserkovskii, B. P.; Frank-Kamenetskii,
M. D.; Egholm, M. E.; Buchardt, O.; Berg, R. H.; Nielsen, P.
E. Proc. Natl. Acad. Sci. USA 1993, 90, 1667.
23. Almarsson, O.; Bruice, T. C.; Kerr, J.; Zuckermann, R. N.
Proc. Natl. Acad. Sci. USA 1993, 90, 7518.
7. Egholm, M.; Buchardt, O.; Nielsen, P. E.; Berg, R. H. J.
Am. Chem. Soc. 1992, 114, 1895.
24. Hyrup, B.; Egholm, M.; Buchardt, O.; Nielsen, P. E.
Bioorg. Med. Chem. Lett. 1996, 6, 1083.
8. Egholm, M.; Nielsen, P. E.; Buchardt, O.; Berg, R. H.; J.
Am. Chem. Soc. 1992, 114, 9677±9678.
25. Lagri€oul, P.-H.; Egholm, M.; Nielsen, P. E.; Berg, R. H.;
Buchardt, O. Bioorg. Med. Chem. Lett. 1994, 4, 1081.
9. Egholm, M.; Buchardt, O.; Christensen, L.; Behrens, C.;
Freier, S. M.; Driver, D. A.; Berg, R. H.; Kim, S. K.; Norden,
B.; Nielsen, P. E. Nature 1993, 365, 566.
26. Krotz, A. H.; Buchardt, O.; Nielsen, P. E. Tetrahedron
Lett. 1995, 36, 6937.
27. Krotz, A. H.; Buchardt, O.; Nielsen, P. E. Tetrahedron
Lett. 1995, 36, 6941.
10. Dueholm, K. L.; Egholm, M.; Behrens, C.; Christensen, L.;
Hansen, H. F.; Vulpius, T.; Petersen, K. H.; Berg, R. H.; Niel-
sen, P. E.; Buchardt, O. J. Org. Chem. 1994, 59, 5767.
28. Loudon, G. M.; Radhakrishna, A. S.; Almond, M. R.;
Blodgett, J. K.; Boutin, R. H. J. Org. Chem. 1984, 49, 4272.
11. Christensen, L.; Fitzpatrick, R.; Gildea, B.; Petersen, K.
H.; Hansen, H. F.; Koch, T.; Egholm, M.; Buchardt, O.; Niel-
sen, P. E.; Coull, J.; Berg, R. H. J. Peptide Sci. 1994, 3, 175.
29. Loudon, G. M.; Almond, M. R.; Jacob, J. N. J. Am. Chem.
Soc. 1981, 103, 4508.
30. Hyrup, B.; Egholm, M.; Rolland, M.; Nielsen, P. E.; Berg, R.
H.; Buchardt, O. J. Chem. Soc. Chem. Commun. 1993, 518.
12. Thomson, S. A.; Josey, J. A.; Cadilla, R.; Gaul, M. D.;
Hassman, F. C.; Luzzio, M. J; Pipe, A. J.; Reed, K. L.; Ricca,
D. J.; Wiethe, R. W.; Noble, S. A. Tetrahedron 1995, 51, 6179.
31. Hyrup, B.; Egholm, M.; Nielsen, P. E.; Wittung, P.; Nor-
den, B.; Buchardt, O. J. Amer. Chem. Soc. 1994, 116, 7964.
13. Kim, S. K.; Nielsen, P. E.; Egholm, M.; Buchardt, O.;
Berg, R. H.; Norden, B. J. Am. Chem. Soc. 1993, 115, 6477.
32. Eriksson, M.; Nielsen, P. E. Nature Struct. Biol. 1996, 3,
410±413.
14. Betts, L; Josey, J. A.; Veal, J. M.; Jordan, S. R. Science
1995, 270, 1838.
33. Brown, S. C.; Thomson, S. A.; Veal, J. M.; Davis, D. G.
Science 1994, 265, 777±780.
15. Nielsen, P. E.; Egholm, M.; Buchardt, O. Gene, 1994, 149,
139.
34. Rasmussen, H.; Kastrup, J. S.; Nielsen, J. N.; Nielsen, J.
M.; Nielsen, P. E. Nature Structural Biology 1997, 4, 98.
16. Nielsen, P. E.; Egholm, M.; Berg, R. H.; Buchardt, O.
Anti-Cancer Drug Design 1993, 8, 53±63; 11, 384.
35. Blessing, R. H. Cryst. Rev. 1987, 1, 3.
17. Hanvey, J. C.; Pe€er, N. J.; Bisi, J. E.; Thomson, S. A.;
Cadilla, R.; Josey, J. A.; Ricca, D. J.; Hassman, C. F.; Bonham,
36. Sheldrick, G. M. Program for the re®nement of crystal
structures. University of Gottingen, Germany, 1993.